A hospital on Mars: surgery, pharmacy and critical care without terrestrial evacuation

The hospital begins with a brutal question: what must be possible before it is too late?

On Earth, a small facility can transfer a difficult case to a larger hospital. Mars removes that escape valve. Appendicitis, major trauma, haemorrhage, severe infection, fracture or obstetric emergency may evolve faster than detailed help can arrive from Earth and infinitely faster than physical evacuation. Local capability must therefore be designed around events that cannot wait.

That does not mean flying a miniature copy of every terrestrial specialty. Mass, volume, energy and staffing force a focus on cross-cutting functions: versatile ultrasound, compact laboratory capability, physiological monitoring, imaging when clinically justified, sterilization, anesthesia, ventilation, surgery and dentistry.

An inventory is not a capability. Equipment, trained staff, drugs, consumables, power and procedures must exist at the same time. A ventilator without oxygen or an ultrasound system without a working probe cannot be counted as care capacity.

The real inventory is a list of functions, not a list of objects

A terrestrial hospital can order a part, transfer a patient, call a specialist and borrow blood from a regional system. On Mars, medical capability has to be described differently: which functions can be completed end-to-end with the people, instruments, consumables, drugs, power and time actually available? An ultrasound machine without a working probe is not a capability. A ventilator without a clean circuit, oxygen, battery and trained operator is not a capability. A surgeon without sterilization, anesthesia and postoperative monitoring is not a capability.

This way of counting leads naturally to a capability matrix. For each priority problem — trauma, hemorrhage, infection, dental pain, dehydration, anaphylaxis, burns, cardiac problems, abdominal emergencies and possible obstetric complications — the settlement describes detection, stabilization, diagnosis, treatment, monitoring and recovery. An empty cell becomes an architecture decision: accept the risk, import a technology, train someone, add inventory, or design a different procedure.

No rapid evacuation: the threshold for local treatment changes completely

ISS medical planning still benefits from a nearby Earth and from return scenarios. Mars removes that safety valve. Even if a return vehicle exists, orbital windows and travel time make Earth useless as an emergency department. The medical system therefore has to move toward Earth-independent medical operations: not absolute independence from terrestrial knowledge, but the ability to make and execute time-critical decisions without waiting for a distant expert to intervene.

That constraint changes training as well. The best physician in the settlement may be the patient. Skills must therefore be distributed: several people able to stabilize an airway, control bleeding, perform basic ultrasound, establish vascular access, manage emergency medications and follow resuscitation procedures. Human redundancy becomes as important as hardware redundancy.

Surgery, anesthesia and intensive care depend on an invisible factory

Surgery requires far more than a surgeon: cleanable space, sterile instruments, anesthesia, monitoring, ventilation, fluids, pain control, sutures, resuscitation and postoperative care. Failure of a sterilizer or pump can remove more clinical capability than failure of a glamorous scanner.

Blood management is a special challenge. Blood products have storage and compatibility constraints, and a small community cannot maintain the inventory of a terrestrial city. Typing, potential donors, preservation strategies and alternatives become part of the architecture and require strict ethical rules.

Medical equipment must itself be repairable. Sensors, batteries, seals, software and pumps need inspection and spares. Biomedical maintenance is part of healthcare; restoring a ventilator may become a clinical emergency.

The operating room is a fluid, power and sterility infrastructure

Surgery consumes much more than instruments. It uses lighting, suction, oxygen, ventilation, medication, fluids, monitoring, sterile barriers, waste handling, water, heat rejection and qualified staff time. Even a relatively modest procedure may occupy several skilled people for hours. In a small settlement, the patient and the medical team therefore disappear from general operational capacity at the same time.

Anesthesia is especially systemic. A drug only makes sense if ventilation, oxygenation, circulation and temperature can be monitored, the airway can be managed and adverse events can be treated. Some agents create environmental or storage constraints. The medical ECLSS therefore cannot be designed independently of the habitat: reduced oxygen production, a power fault or a contamination event can turn a surgical problem into a habitat-survival problem.

Blood: scarce, perishable and biologically difficult to replace

Blood illustrates the logistics problem better than almost any other medical resource. A tiny population does not have an unlimited compatible donor pool. Blood products have different storage conditions and shelf lives, and transfusion requires identification, compatibility, procedure and monitoring. A settlement may therefore combine hemorrhage prevention, rapid mechanical control, appropriate blood-conservation techniques, limited stored products and carefully governed emergency donor protocols.

Demography matters. A four-person crew cannot rely on the immuno-hematologic diversity of a city. A population of one hundred or one thousand expands the donor pool, but also increases demand. Blood banking gradually becomes a laboratory and public-health function, which is another example of why settlement growth changes the category of medicine rather than simply scaling the first-aid kit.

Robotic surgery: useful assistance, dangerous as a magical promise

Robotic surgery demonstrations in microgravity have shown that selected manipulations can be teleoperated and studied remotely. Mars creates a hard boundary: delay prevents continuous teleoperation from Earth. Robotics can stabilize an instrument, guide a motion, reproduce a trajectory or assist a local operator; it does not remove the need for local caregivers or for a fallback plan when the robot fails.

Pharmacy: expiration dates become mission variables

A pharmacy is not a fixed mass. Drugs age, some require controlled temperature, and use is unpredictable. Each lot should be tracked by indication, substitute, criticality, expiration, consumption and resupply time.

Local production may begin with modest capabilities: simple solutions, compounding, excipients, packaging or selected molecules compatible with local chemistry. Reproducing a modern pharmacopoeia would require fine chemistry, quality control and regulatory capability far beyond basic ISRU.

The likely strategy is a portfolio: stable prepositioned medicines, stock rotation by cargo, generous reserves of critical products, therapeutic alternatives and limited local manufacture. A thousand-person city can justify capabilities that a four-person base cannot.

From terrestrial inventory to astropharmacy

A medicine sent to Mars may spend months in terrestrial storage, months in transit and years inside a habitat exposed to radiation and controlled but imperfect thermal conditions. Shelf life is not merely a printed date; it is a property of formulation, packaging, temperature, humidity, light and radiation. A Martian pharmacy must know not only how much inventory remains but how much confidence remains in its quality.

The first improvement is stock rotation and therapeutic redundancy. If two medicines can address the same indication through different formulations or mechanisms, loss of one does not immediately eliminate treatment. On the other hand, storing fifty rarely used variants consumes mass and volume. Selection must be driven by risk, expected frequency, severity and substitution options.

Local production should start with simple dosage forms

Local pharmaceutical manufacturing probably does not begin with complete synthesis of complex molecules. It may begin with pharmaceutical-quality water, simple solutions, repackaging, dilution, compounding and selected consumables. Every step still requires feedstock purity, dose verification, sterility where required and analytical control. A medicine that is “made” without evidence of potency or contamination control is not reliable medical capability.

Biotechnology and chemical synthesis may eventually expand autonomy, but the difficulty is not only the reaction. Precursors, solvents, catalysts, equipment, documentation and a local pharmacopoeia are needed. Pharmacy therefore joins the industrial chapter: therapeutic autonomy depends on a quality chain comparable with that of a critical mechanical part.

Antibiotics connect inventory with microbial ecology

A small community may be tempted to use scarce antibiotics defensively. That behavior can select resistance. Local microbiology should therefore support prescribing when possible: identify organisms, tailor duration, monitor infection and protect last-line drugs. On Mars, antimicrobial resistance is not a distant public-health abstraction; it can make an irreplaceable part of the pharmacy ineffective.

A mission pharmacy is a portfolio of chemical risks

A medicine is not simply “present” or “absent.” It has a formulation, concentration, package, storage temperature, moisture sensitivity, sometimes a cold-chain requirement, and a stability profile that may change with radiation and time. The Martian challenge is therefore to know that a tablet, injectable or solution still has the expected quality when it is needed, potentially years after it was packed.

Initial selection should consider not only clinical usefulness but logistical robustness: number of indications covered, stability, mass, volume, dose flexibility, therapeutic alternatives and consequence of absence. Inventory must be tracked by lot and expiry, with rotation and consumption priorities. Prescribing therefore becomes partly a logistics operation: using a rare drug today may reduce the ability to manage a future emergency.

Inventory arithmetic: mass is only the beginning

Suppose a settlement stores 20,000 administration units across multiple therapeutic classes, a figure chosen only to illustrate the method. If the average packaged mass is 2 grams per unit:

20,000 × 2 g = 40,000 g = 40 kg.

Forty kilograms says very little about resilience. If half of the inventory expires in the same year, if an essential class has only one therapeutic option, or if critical products depend on one refrigerator, the system remains fragile. Planning therefore needs temporal coverage, therapeutic redundancy, expiry distribution, cold-chain dependence and resupply interval, not merely total mass.

Astropharmacy: local production without confusing synthesis with a qualified medicine

Recent NASA work explores on-demand production of small pharmaceutical quantities during deep-space missions, including synthetic-biology platforms. The idea is powerful: instead of shipping every molecule for the entire life of a settlement, import equipment and feedstocks that can produce selected compounds. But the distance between “a molecule was produced” and “a medicine can be safely administered” is enormous.

The product has to be identified, purity and concentration verified, contaminants or by-products removed, a suitable dosage form prepared, sterility assured when required, the batch documented and stability understood. Martian pharmacy therefore connects directly to industrial chemistry, biotechnology, metrology and quality assurance. Local production will probably begin with a small number of high-value products and well-controlled processes rather than a miniature copy of Earth’s pharmaceutical industry.

Telemedicine: delayed expertise, never remote control of an emergency

Earth remains a huge medical resource, able to gather specialists and review difficult cases. But latency prevents real-time remote control of resuscitation or surgery. Local teams must be able to act, stabilize and formulate structured questions before Earth answers.

Digital tools can help: offline medical records, procedure libraries, simulation, guided ultrasound and decision support. They must be versioned, auditable and usable without an Earth server. A critical medical application that requires continuous terrestrial connectivity is a poor Mars design.

TRISH and HERMES illustrate the value of health data that follows the explorer and remains useful under limited connectivity. At settlement scale, that becomes a local health information system with privacy, emergency access, backup and epidemiological analysis.

Martian telemedicine will look more like a documented consultation than a video call

As delay grows, the best interaction is no longer “watch this live image and tell me what to do.” A consultation package sent to Earth should be self-contained: summary, vital signs, evolution, images, laboratory results, treatment already given and explicit questions. Earth specialists can then return a structured assessment. Asynchronous communications can therefore improve local documentation discipline.

Solar conjunction or network failure also requires periods with no outside expertise. Every critical protocol needs an offline form with escalation levels, limits and decision points. The medical centre must be able to operate like an isolated ship and later reconcile its records with Earth when communications return.

From four to one thousand people: medicine changes category

With four people, versatility dominates. Everyone needs emergency skills and the only physician may become the patient. Human redundancy is as important as hardware redundancy.

At twenty, a permanent clinic, deeper pharmacy and cross-trained skills are possible. At one hundred, a small hospital can separate consultation, isolation, laboratory and procedures. At one thousand, specialties, rehabilitation, obstetrics, pediatrics and professional biomedical maintenance become necessary.

Scale also changes epidemiology. A single accident dominates a tiny crew; chronic disease, prevention, vaccination, mental health and ageing become continuing public-health work in a city. The medical system stops being a mission kit and becomes a civic institution.

Four, twenty, one hundred, one thousand: four different medical architectures

Four residents. Medicine resembles an extreme expedition: one person with advanced medical training, several cross-trained crewmembers, ultrasound, a narrow laboratory, pre-positioned pharmacy and aggressive prevention. The aim is to stabilize, manage plausible events and avoid losing the system if the single expert becomes the patient.

Twenty residents. Skills can be distributed. On-call coverage becomes possible, dentistry has to be real, the laboratory expands, infection isolation becomes less improvised and rehabilitation matters more. Rare procedures must be rehearsed so that competence does not decay.

One hundred residents. Medical selection can no longer remove most chronic disease. The settlement has to manage long-term conditions, possible pregnancy, aging pioneers, industrial accidents and occupational health. A small operating room, structured pharmacy, broader laboratory and specialized staff become rational.

One thousand residents. The settlement requires a real hospital, although still compact by terrestrial standards: critical care, surgery, possible maternity, pediatrics, rehabilitation, mental health, public health, dentistry, laboratory, pharmacy, sterilization and biomedical maintenance. It must also handle peaks such as fire, transport accidents, decompression, food contamination or industrial events that produce several casualties at once.

Medical skills are spare parts too

A rare skill can disappear when one person is injured, dies or is exhausted. The settlement therefore needs a competency inventory much like its inventory of filters and bearings: at least two or three people able to perform each critical function, local documentation, regular simulation and continuous teaching. In the long term, schools and professional training must be capable of producing the next generation of caregivers without relying on immigration from Earth.